Fuzzface issues!

[Flynn]

Hello all,

I breadboarded this and got some fuzziness right away.  But, its not right.  Not very rich, and then as the sustain dies out it sounds like the battery going out on a pedal (abrupt and sputtering).


I used this schematic:



Positive Ground.

I am using higher value pots for most everything (500kb for output, 10k for bias/sundial, 5k for fuzz) but the rest of the part values are stock to the above schematic.


Wallwart power supply:  9.22v

Transistors: 2N527's. 

Tested using RG's tester using 2.418K and 2.18M resistance.  Values show "before / after button push"

1st 2N527
100mv / .665v


2nd 2N527
100mv / .728v

My cheap multimeter said their Hfe's were 104 and 116.


Q1 Readings:

B:  -9.10v
C:  -8.53v
E:  -9.22v

Q2 Readings:
B:  -8.53v
C:  -8.65v
E:  -8.66v

No matter WHAT i have tried I have not been able to get the Q2 Collector anywhere less than around this 8.6 volts.   I have tried changing the R1 (100k) resistor value up and down, adding a pot in parallel to the fuzz pot, and a bunch of other attempts and nothing has done it.  I have poked and re-plugged wires, etc.

I have triple checked cap orientation and the schemo as a whole.

I also put a 2N3906 in the Q2 spot and no difference in voltage reading on the collector (and not much on the sound, either, oddly enough).

Also not sure if these transistors' readings are saying they're cool for this project or not. 

I'm baffled and must now turn to those more knowledgeable than myself!

[Electric Warrior]

Something's wrong here. Q1's emitter isn't connecting with ground.

[Flynn]

hey man, thanks for the input.  i have the leg of the thing literally plugged into the ground plane of the breadboard. i just checked it again and it's connecting.  so nutz.

[Clint Eastwood]

R4 470k?? that would be about 1000 times too much. Try 470 ohms.

[thermionix]

That schematic is faulty.  I think it's supposed to be the Fulltone (not sure).  But the 470k resistor up top should be 470 ohms, or something close to that.  Major difference.  The 100k on Q1e isn't right either.  100 ohms would usually work, but original FFs went straight to ground.

Edit:  Beat by Clint, but posting anyway to back him up.  (As if Clint Eastwood ever needed backup)

[Electric Warrior]

Q1E does go straight to ground. The 100k connects with the base.

[Mr.Kite]

Also the 5k Fuzz pot you're using can mess with the bias, but first replace the 470k with a 470R...

[Flynn]

Q1 E goes direct to ground.

Will putting a 1k on the fuzz pot help with the bias even though its more limited than a 5k?

Put a 470 in there instead of the 470k - circuit started screaming, no love. 

Wondering if it's a mistake in the schemo or part of an overall circuit change.  Or, since the 2n3906 didn't make much of a change, a problem with the Q2 section?


Thanks so much for all the feedback.

[thermionix]

Q1E does go straight to ground. The 100k connects with the base.

Oops!  My bad!  I saw that wrong, somehow.

[Flynn]

hey all good!   and where can i get that t-shirt!!!!!  ha! :-)

[thermionix]

I think it's supposed to be the Fulltone (not sure).

Also, obviously an AM Sunface, not a Fulltone.

Just ignore me.

[Jolly Jimmy]

Q1 Readings:

B:  -9.10v
C:  -8.53v
E:  -9.22v

Q2 Readings:
B:  -8.53v
C:  -8.65v
E:  -8.66v

I'm guessing the -9.22v reading on Q1's emitter is just the total supply voltage and that you have your power rails mixed up somehow. Make sure ground is connected to the + side of the power supply and that it's the reference point from which you take your readings. Q1E should read 0V.

[Slowpoke101]

Try removing the ground connections on the Sundial 5K & Trimmer 50K pots.
Change R4 for 1K2 (anywhere between 1K to 3K3).
See what happens. Good luck.

[lion]

....Put a 470 in there instead of the 470k - circuit started screaming, no love.

Try again

The schematic you are showings seems to be made by Benoit M/Coda Effects - and he states here http://www.coda-effects.com/p/circuit-analysis-fuzz-face.html that R4 470 kilo-ohms is incorrect (scroll down to the bottom).

Erik

[Juan Wayne]

You know, sometimes you're better off starting all over again, fresh head with a fresh breadboard. Oh, and definitely 470Ohms instead of 470k, in case we didn't mention that enough.

[Flynn]

Hey - i wanted to report back:

I listened and attempted all the suggestions.  Still wasn't working for me.

I then just cleared the board and started again, per Juan Wayne, and bingo.

If I had to guess I'd say that my un-tinned wiper lead of the Sundial didn't make a good connection in the breadboard.

Sounds seriously good now.  Not sure if anyone can look at my Transistor data and tell me what their gains are (per R.G.'s specs), but it sure does sound good.

The last thing i'm futzin with now is whether to use a 5k or 10k on the Sundial (bias).  The 5k in series with the 2.2k resistor doesn't add up to the original 8.2k value.  With the 5k i get 6.78v (max) on Q2 Collector, with the 10k its 7.16v and seems a bit brighter/edgier. 

Can't thank you guys enough for the suggestions, i really do appreciate it.

Just can't believe none of you mentioned i should use a 470 instead of a 470k!!!!   

[Juan Wayne]

Why would you need a 470Ohm? I don't get it... Anyways, good thing it worked out!

I too would go for 10k on the bias, at least try it and see if the whole range is usable before commiting to the definitive circuit. I think mine ended up with 20k and no limiting resistance. I use the whole range.

[Flynn]

Thanks Juan Wayne, just tried a 50k and bypassed the 2.2k - it goes into feedback, so that's staying!  Will put it on a switch tho so i can flip that 2.2k in.

I tried a few diff values there - 1.5k, 2.2k and 3.3k and I must admit it seemed to roll off the low-end differently than simply dialing the Sundial back to match the same voltage.  Does that make sense? Maybe I should use one of these 4-way rotary switches i have.

[Clint Eastwood]

Sounds seriously good now.  Not sure if anyone can look at my Transistor data and tell me what their gains are (per R.G.'s specs), but it sure does sound good.

If I am not mistaken, You have to substract the first voltage value from the second and then multiply it by 100. For Q1 this gives a Hfe of (665-100)*100 = 57 and for Q2 (728-100)*100 = 73.  Your multimeter gives you fake news, probably because it denies leakage.
These values for Hfe are obviously  lower than what is commonly regarded as appropriate for a fuzz face, so it might sound even better with higher gain devices, but I am very happy for you to now have a working circuit that sounds good!

[Flynn]

Thanks Clint Eastwood.  I will be returning to my local shop to return some faulty AC128's and will pick up some more of these 2N527's and test them out. 

I am installing transistor mounts so i can swap things out easily and not screw up the transistor with soldering heat.

With these transistors and the 50k Sundial it definitely hits the extreme.  Wild, feedbacky, compressed. 

Would higher gain ones mean it would break up sooner or have a tonal difference or anything?